Section Gears Up for 2010 JSM
Edited by Page Moore, Biometrics Section Publications Officer
The Biometrics Section is proud to sponsor one continuing education course and four invited sessions at this year’s Joint Statistical Meetings.
Continuing Education Course: Regression Modeling Strategies—
Frank Harrell Jr. will give a one‑day short course on regression modeling strategies. He is professor and chair, department of biostatistics, Vanderbilt University School of Medicine, in Nashville, Tennessee. Regression models are frequently used to develop diagnostic, prognostic, and health resource utilization models in clinical, health services, outcomes, pharmacoeconomic, and epidemiologic research, and in a multitude of nonhealth-related areas. Regression models are also used to adjust for patient heterogeneity in randomized clinical trials, to obtain tests that are more powerful and valid than unadjusted treatment comparisons.
Models must be flexible enough to fit nonlinear and nonadditive relationships. However, unless the sample size is enormous, the approach to modeling must avoid common problems with data mining or data dredging that result in overfitting and a failure of the predictive model to validate on new subjects.
All standard regression models have assumptions that must be verified for the model to have power to test hypotheses and to predict accurately. Of the principal assumptions (linearity, additivity, distributional), this short course will emphasize methods for assessing the first two. Practical but powerful tools are presented for validating model assumptions and presenting model results. This course provides methods for estimating the shape of the relationship between predictors and response.
The first part of the course presents the following elements of multivariable predictive modeling for a single response variable: using regression splines to relax linearity assumptions, perils of variable selection and overfitting, where to spend degrees of freedom, shrinkage, imputation of missing data, data reduction, and interaction surfaces. Then a default overall modeling strategy will be described, with an eye toward “safe data mining.” This is followed by methods for graphically understanding models (e.g., using nomograms) and using resampling to estimate a model’s likely performance on new data.
Participants should have a good working knowledge of multiple regression and should consider reading the following articles in advance:
In addition, the Biometrics Section is sponsoring an exciting program of invited sessions and talks spanning a broad range of topics in biostatistics. The titles and organizers of the invited sessions are:
- “Statistical Evaluation of Markers Used to Select Treatment,” Margaret Pepe, University of Washington
- “Study Design and Statistical Analysis Challenges in Women’s Health Studies,” Marcia Ciol, University of Washington
- “Evaluation of Risk Prediction,” Shulamith Gross, Baruch College
- “Getting More from Genome-Wide Association Studies,” Mitchell Gail, National Cancer Institute
The section thanks Jerry Heatley, our JSM Continuing Education Chair, and Hormuzd Katki, our JSM Program Chair, for organizing these courses and sessions. Check the online program at the 2010 JSM web site for updates on locations and times.